Computer Sciences Commons^™

A Convex Optimization Algorithm For Sparse Representation And Applications In Classification Problems, Reinaldo Sanchez Arias

Open Access Theses & Dissertations

In pattern recognition and machine learning, a classification problem refers to finding an algorithm for assigning a given input data into one of several categories. Many natural signals are sparse or compressible in the sense that they have short representations when expressed in a suitable basis. Motivated by the recent successful development of algorithms for sparse signal recovery, we apply the selective nature of sparse representation to perform classification. Any test sample is represented in an overcomplete dictionary with the training sample as base elements. A given test sample can be expressed as a linear combination of only those training …

Design And Practical Application Of An Innovative, Pneumatically Latched Valve, Cheng Y. Lin, Jennifer G. Michaeli, Nathan J. Luetke

Engineering Technology Faculty Publications

This article explains the design and fabrication of an innovative five-port, two-position (5/2), pneumatically latched valve. The initial application to repeatedly and automatically lift a mechanical cover or gate was presented as part of an undergraduate student project in the Mechanical Engineering Technology Program's Automation and Controls course in the Department of Engineering Technology at Old Dominion University. Valve operation, including self-latching, utilizes mechanical means only. The valve does not require any electrical power, electronic sensors, or controller, which makes it an energy-saving device. For light-duty operation, a bellows foot-air pump is used for air supply. A single activation of …

Orientation Invariant Ecg-Based Stethoscope Tracking For Heart Auscultation Training On Augmented Standardized Patients, Nahom Kidane, Salim Chemlal, Jiang Li, Frederic D. Mckenzie, Tom Hubbard

Computational Modeling & Simulation Engineering Faculty Publications

Auscultation, the act of listening to the heart and lung sounds, can reveal substantial information about patients’ health and other cardiac-related problems; therefore, competent training can be a key for accurate and reliable diagnosis. Standardized patients (SPs), who are healthy individuals trained to portray real patients, have been extensively used for such training and other medical teaching techniques; however, the range of symptoms and conditions they can simulate remains limited since they are only patient actors. In this work, we describe a novel tracking method for placing virtual symptoms in correct auscultation areas based on recorded ECG signals with various …

Integration Of Multispectral Face Recognition And Multi-Ptz Camera Automated Surveillance For Security Applications, Chung-Hao Chen, Yi Yao, Hong Chang, Andreas Koschan, Mongi Abidi

Electrical & Computer Engineering Faculty Publications

Due to increasing security concerns, a complete security system should consist of two major components, a computer-based face-recognition system and a real-time automated video surveillance system. A computer-based face-recognition system can be used in gate access control for identity authentication. In recent studies, multispectral imaging and fusion of multispectral narrow-band images in the visible spectrum have been employed and proven to enhance the recognition performance over conventional broad-band images, especially when the illumination changes. Thus, we present an automated method that specifies the optimal spectral ranges under the given illumination. Experimental results verify the consistent performance of our algorithm via …

A Webquest For The Instruction Of Appropriate Online Behavior, Susan Heilig

Graduate Research Papers

The requirements of the Children's Internet Protection Act (CIPA) and the expectations of the Iowa Department of Education's (2012) Iowa Core Curriculum 21st Century Skills increase the importance of having an organized collection of resources to teach Internet safety. These requirements and the literature reviewed confirmed the importance of preparing students to use the Internet safely and ethically and be productive digital citizens. While the teachers in the Calamus-Wheatland School District were already instructing students in appropriate online behavior, there wasn't an organized Internet Safety resource that could be used effectively within the time limits of the scheduled computer classes. …

Adaptive Event-Triggered Control Of A Uncertain Linear Discrete Time System Using Measured Input And Output Data, Avimanyu Sahoo, Hao Xu, S. Jagannathan

Electrical and Computer Engineering Faculty Research & Creative Works

In this paper, an adaptive model-based event-triggered control of an uncertain linear discrete time system is developed. Measured input and output vectors and their history are utilized to express the unknown linear discrete-time system as an autoregressive Markov representation (ARMarkov). A novel adaptive model in the form of AR Markov is proposed and an update law is derived in order to estimate parameters of the ARMarkov model at triggered instants unlike periodic updates in standard adaptive control. Lyapunov method is used to derive the event trigger condition, prove boundedness of the parameter vector and asymptotic convergence of the outputs and …

Neural Network-Based Adaptive Event-Triggered Control Of Affine Nonlinear Discrete Time Systems With Unknown Internal Dynamics, Avimanyu Sahoo, Hao Xu, S. Jagannathan

Electrical and Computer Engineering Faculty Research & Creative Works

In this paper, the design of a neural network (NN) based adaptive model-based event-triggered control of an uncertain single input single output (SISO) nonlinear discrete time system in affine form is presented. The controller uses an adaptive estimator consisting of a single-layer NN not only to approximate the internal dynamics of an affine nonlinear discrete-time system but also to provide an estimate of the state vector during inter event interval. The NN weights of the adaptive NN estimator are tuned in a aperiodic manner at the event trigger instants unlike periodic updates in standard adaptive neural network (NN) control. A …

A Latent Dirichlet Allocation/N-Gram Composite Language Model, Raymond Daniel Kulhanek

Browse all Theses and Dissertations

I present a composite language model in which an n-gram language model is integrated with the Latent Dirichlet Allocation topic clustering model. I also describe a parallel architecture that allows this model to be trained over large corpora and present experimental results that show how the composite model compares to a standard n-gram model over corpora of varying size.

A Large Scale Distributed Syntactic, Semantic And Lexical Language Model For Machine Translation, Ming Tan

Browse all Theses and Dissertations

The n-gram model is the most widely used language model (LM) in statistical machine translation system, due to its simplicity and scalability. However, it only encodes the local lexical relation between adjacent words and clearly ignores the rich syntactic and semantic structures of the natural languages. Attempting to increase the order of an n-gram to describe longer range dependencies in natural language immediately runs into the curse of dimensionality. Although previous researches tried to increase the order of n-gram on a large corpus, they did not see obvious improvement beyond 6-gram. Meanwhile, other LMs, such as syntactic language models and …

A Justification For Semantic Training In Data Curation Frameworks Development, Xiaogang Ma, Benjamin D. Branch, Kristin Wegner

Libraries Faculty and Staff Presentations

In the complex data curation activities involving proper data access, data use optimization and data rescue, opportunities exist where underlying skills in semantics may play a crucial role in data curation professionals ranging from data scientists, to informaticists, to librarians. Here, We provide a conceptualization of semantics use in the education data curation framework (EDCF) (Fig. 1) [1] under development by Purdue University and endorsed by the GLOBE program [2] for further development and application. Our work shows that a comprehensive data science training includes both spatial and non-spatial data, where both categories are promoted by standard efforts of organizations …

Visual Exploration And Information Analytics Of High-Dimensional Medical Images, Darshan Pai

Wayne State University Dissertations

Data visualization has transformed how we analyze increasingly large and complex data sets. Advanced visual tools logically represent data in a way that communicates the most important information inherent within it and culminate the analysis with an insightful conclusion. Automated analysis disciplines - such as data mining, machine learning, and statistics - have traditionally been the most dominant fields for data analysis. It has been complemented with a near-ubiquitous adoption of specialized hardware and software environments that handle the storage, retrieval, and pre- and postprocessing of digital data. The addition of interactive visualization tools allows an active human participant in …

Combinatorial Auction-Based Virtual Machine Provisioning And Allocation In Clouds, Sharrukh Zaman

Wayne State University Dissertations

Current cloud providers use fixed-price based mechanisms to allocate Virtual Machine (VM) instances to their users. But economic theory states that when there are large amount of resources to be allocated to large number of users, auctions are the most efficient allocation mechanisms. Auctions achieve efficiency of allocation and also maximize the providers' revenue, which a fixed-price based mechanism is unable to do. We argue that combinatorial auctions are best suited for the problem of VM provisioning and allocation in clouds, since they provide the users with the most flexible way to express their requirements. In combinatorial auctions, users bid …

Fault Management For Service-Oriented Systems, Amal Alhosban

Wayne State University Dissertations

Service Oriented Architectures (SOAs) enable the automatic creation of business applications from independently developed and deployed Web services. As Web services are inherently unreliable,

how to deliver reliable Web services composition over unreliable Web services is a significant and challenging problem. The process requires monitoring the system's behavior, determining when and why faults occur, and then applying fault prevention/recovery mechanisms to minimize the impact and/or recover from these faults. However, it is hard to apply a non-distributed management

approach to SOA, since a manager needs to communicate with the different components through authentications. In SOA, a business process can terminate …

Supporting Text Retrieval Query Formulation In Software Engineering, Sonia Cristina Haiduc

Wayne State University Dissertations

The text found in software artifacts captures important information. Text Retrieval (TR) techniques have been successfully used to leverage this information. Despite their advantages, the success of TR techniques strongly depends on the textual queries given as input. When poorly chosen queries are used, developers can waste time investigating irrelevant results.

The quality of a query indicates the relevance of the results returned by TR in response to the query and can give an indication if the results are worth investigating or a reformulation of the query should be sought instead. Knowing the quality of the query could lead to …

Opacity Of Discrete Event Systems: Analysis And Control, Majed Mohamed Ben Kalefa

Wayne State University Dissertations

The exchange of sensitive information in many systems over a network can be manipulated

by unauthorized access. Opacity is a property to investigate security and

privacy problems in such systems. Opacity characterizes whether a secret information

of a system can be inferred by an unauthorized user. One approach to verify security

and privacy properties using opacity problem is to model the system that may leak confidential

information as a discrete event system. The problem that has not investigated

intensively is the enforcement of opacity properties by supervisory control. In other

words, constructing a minimally restrictive supervisor to limit the system's …

The Role Of Test Administrator And Error, Michael Edward Brockly

Open Access Theses

This study created a framework to quantify and mitigate the amount of error that test administrators introduced to a biometric system during data collection. Prior research has focused only on the subject and the errors they make when interacting with biometric systems, while ignoring the test administrator. This study used a longitudinal data collection, focusing on demographics in government identification forms such as driver's licenses, fingerprint metadata such a moisture and skin temperature, and face image compliance to an ISO best practice standard. Error was quantified from the first visit and baseline test administrator error rates were measured. Additional training, …

Integrating The Sap Software Into The Manufacturing Engineering Curriculum, Lin Qian

Open Access Theses

The concept of Enterprise Resource Planning (ERP) was developed to integrate data efficiently and eliminate redundancy within organizations. Worldwide industries rely on ERP systems to keep a competitive advantage in world market place. Besides, cross-functional knowledge is required in industry. It is critical that schools integrate ERP concepts and methods in to academic curricula for Technology and Engineering School.

This thesis examined the effectiveness of a simulation game as a method for teaching for ERP systems in Technology and Engineering. The quasi-experimental design was used in this study to determine if the SAP simulation game had an impact on students' …

Interactive Multivariate Data Exploration For Risk-Based Decision Making, Silvia Oliveros Torres

Open Access Theses

Risk-based decision making is a data-driven process used to gather data about outcomes, analyze different scenarios, and deliver informed decisions to mitigate risk.

An interactive visual analytics system can help derive insights from large amounts of data and facilitate the risk management process thereby providing a suitable solution to examine different decision factors. This work introduces two separate systems that

incorporate visual analytics techniques to help the users in the tasks of identifying patterns, finding correlations, designing mitigation strategies, and assisting in the long term planning and assessment process. The first system looks at the National Health

and Nutrition Examination …

An Analysis Of Underlying Competencies And Computer And Information Technology Learning Objectives For Business Analysis, Ryan Thomas Quigley

Open Access Theses

This research examines whether the Computer and Information Technology (CIT) department at Purdue University should develop a business analyst concentration. The differences between system and business analysts, evolution of the business analyst profession, job demand and trends, and applicable model curricula were explored to support this research. Review of relevant literature regarding the topics suggested that a business analyst concentration should be developed. A gap analysis was performed to determine how well selected CIT courses address the skills and competencies required by today's business analysts. The primary finding, as a result of the analysis, was that CIT courses alone are …

Distributed Digital Forensics On Pre-Existing Internal Networks, Jeremiah Jens Nielsen

Open Access Theses

Today's large datasets are a major hindrance on digital investigations and have led to a substantial backlog of media that must be examined. While this media sits idle, its relevant investigation must sit idle inducing investigative time lag. This study created a client/server application architecture that operated on an existing pool of internally networked Windows 7 machines. This distributed digital forensic approach helps to address scalability concerns with other approaches while also being financially feasible. Text search runtimes and match counts were evaluated using several scenarios including a 100 GB image with prefabricated data. When compared to FTK 4.1, a …

Text-Based Phishing Detection Using A Simulation Model, Gilchan Park

Open Access Theses

Phishing is one of the most potentially disruptive actions that can be performed on the Internet. Intellectual property and other pertinent business information could potentially be at risk if a user falls for a phishing attack. The most common way of carrying out a phishing attack is through email. The adversary sends an email with a link to a fraudulent site to lure consumers into divulging their confidential information. While such attacks may be easily identifiable for those well-versed in technology, it may be difficult for the typical Internet user to spot a fraudulent email.

The emphasis of this research …

Simulating Land Use Land Cover Change Using Data Mining And Machine Learning Algorithms, Amin Tayyebi

Open Access Dissertations

The objectives of this dissertation are to: (1) review the breadth and depth of land use land cover (LUCC) issues that are being addressed by the land change science community by discussing how an existing model, Purdue's Land Transformation Model (LTM), has been used to better understand these very important issues; (2) summarize the current state-of-the-art in LUCC modeling in an attempt to provide a context for the advances in LUCC modeling presented here; (3) use a variety of statistical, data mining and machine learning algorithms to model single LUCC transitions in diverse regions of the world (e.g. United States …

Information Measures For Statistical Orbit Determination, Alinda Kenyana Mashiku

Open Access Dissertations

The current Situational Space Awareness (SSA) is faced with a huge task of tracking the increasing number of space objects. The tracking of space objects requires frequent and accurate monitoring for orbit maintenance and collision avoidance using methods for statistical orbit determination. Statistical orbit determination enables us to obtain estimates of the state and the statistical information of its region of uncertainty given by the probability density function (PDF). As even collision events with very low probability are important, accurate prediction of collisions require the representation of the full PDF of the random orbit state. Through representing the full PDF …

Improving Performance By Re-Rating In The Dynamic Estimation Of Rater Reliability, Alexey Tarasov, Sarah Jane Delany, Brian Macnamee

Conference papers

Nowadays crowdsourcing is widely used in supervised machine learning to facilitate the collection of ratings for unlabelled training sets. In order to get good quality results it is worth rejecting results from noisy/unreliable raters, as soon as they are discovered. Many techniques for filtering unreliable raters rely on the presentation of training instances to the raters identified as most accurate to date. Early in the process, the true rater reliabilities are not known and unreliable raters may be used as a result. This paper explores improving the quality of ratings for train- ing instances by performing re-rating. The re-rating relies …

Automated Detection Of Vehicles With Machine Learning, Michael N. Johnstone, Andrew Woodward

Australian Information Security Management Conference

Considering the significant volume of data generated by sensor systems and network hardware which is required to be analysed and intepreted by security analysts, the potential for human error is significant. This error can lead to consequent harm for some systems in the event of an adverse event not being detected. In this paper we compare two machine learning algorithms that can assist in supporting the security function effectively and present results that can be used to select the best algorithm for a specific domain. It is suggested that a naive Bayesian classiifer (NBC) and an artificial neural network (ANN) …

A Matlab Primer In Four Hours With Practical Examples, Jerome Casey

Instructional Guides

No abstract provided.

A Semantics-Based Approach To Machine Perception, Cory Andrew Henson

Browse all Theses and Dissertations

Machine perception can be formalized using semantic web technologies in order to derive abstractions from sensor data using background knowledge on the Web, and efficiently executed on resource-constrained devices. Advances in sensing technology hold the promise to revolutionize our ability to observe and understand the world around us. Yet the gap between observation and understanding is vast. As sensors are becoming more advanced and cost-effective, the result is an avalanche of data of high volume, velocity, and of varied type, leading to the problem of too much data and not enough knowledge (i.e., insights leading to actions). Current estimates predict …

A Methodology For Extracting Human Bodies From Still Images, Athanasios Tsitsoulis

Browse all Theses and Dissertations

Monitoring and surveillance of humans is one of the most prominent applications of today and it is expected to be part of many future aspects of our life, for safety reasons, assisted living and many others. Many efforts have been made towards automatic and robust solutions, but the general problem is very challenging and remains still open. In this PhD dissertation we examine the problem from many perspectives. First, we study the performance of a hardware architecture designed for large-scale surveillance systems. Then, we focus on the general problem of human activity recognition, present an extensive survey of methodologies that …

How Ideas Grow: Critical Mass In The Linear Threshold Model, Hossein Alidaee

Mathematics, Statistics, and Computer Science Honors Projects

We study how ideas spread through a social network using the Linear Threshold Model. Each node i on the complete graph K_n is given a threshold Ɵ_i chosen uniformly at random from (0, 1]. This threshold indicates the fraction of the social network that must be active (or believe the idea) prior to node i becoming active. We start with an activated group of early adopters, called the seed set. Considering various scenarios, we use the probabilistic method to find lower bounds on size of a seed set which guarantees that all nodes become active with high …

Anomalies In Sensor Network Deployments: Analysis, Modeling, And Detection, Giovani Rimon Abuaitah

Browse all Theses and Dissertations

A sensor network serves as a vital source for collecting raw sensory data. Sensor data are later processed, analyzed, visualized, and reasoned over with the help of several decision making tools. A decision making process can be disastrously misled by a small portion of anomalous sensor readings. Therefore, there has been a vast demand for mechanisms that identify and then eliminate such anomalies in order to ensure the quality, integrity, and/or trustworthiness of the raw sensory data before they can even be interpreted.

Prior to identifying anomalies, it is essential to understand the various anomalous behaviors prevalent in a sensor …